Abstract
The authors’ purpose of this study was to evaluate the validity and reliability of the Cardiovascular Symptom Index for Midlife Women in four ethnic/racial groups of midlife women in the U.S. Data from two internet surveys were used for this secondary analysis. In the construct validity test, five factors were extracted among White and Hispanic women, while more than five factors were extracted among Asian and African American women. In the convergent validity test, all ethnic/racial groups showed similar patterns. Adequate internal consistency among was shown. The Cardiovascular Symptom Index for Midlife Women is useful for assessing cardiovascular symptoms in multiethnic/racial women.
Most midlife women could experience cardiovascular symptoms during the menopausal transition (Thurston, Sutton-Tyrrell, Everson-Rose, Hess, & Matthews, 2008) and they may ignore the symptoms because they are dissimilar to what they consider typical cardiovascular symptoms, and they could misunderstand them as discomfort due to estrogen fluctuations during menopause. Because of this, women having heart problems might not recognize their symptoms as cardiac in nature and, therefore, delay seeking medical care and, as a result, have more complications and higher mortality (McSweeney et al., 2003; Rosenfeld, 2001). Thus, it is important to detect the symptoms of cardiovascular disease (CVD) early and accurately in women. Several reseachers reported the idea of ethnic/racial difference among cardiovascular symptoms in women (Eastwood et al., 2013; McSweeney, O’Sullivan, et al., 2010). These facts explained the need to assess the cardiovascular symptoms of multiethnic/racial midlife women accurately.
Measurement of cardiovascular symptom in women
Several instruments have been developed for measuring cardiovascular symptoms such as the Rose Questionnaire (RQ), the Symptoms of Acute Coronary Syndrome Inventory (SACSI), and the McSweeney Acute and Prodromal Myocardial Infarction Symptom Survey (MAPMISS). RQ was developed to facilitate screening for the presence of coronary artery disease and was originally designed for use in men (Rose, McCartney, & Reid, 1977). This questionnaire includes questions on chest pain and intermittent claudication (Rose et al., 1977). However, although this instrument has been used widely with both men and women for epidemiologic studies, some studies have shown that RQ can only predict major coronary events for white men (Oei et al., 2004; Sorlie, Cooper, Schreiner, Rosamond, & Szklo, 1996). It was also reported to be less valid in young women (Harris & Weissfeld, 1991; Nicholson, White, Macfarlane, Brunner, & Marmot, 1999) and in cross-cultural comparison (Fischbacher, Bhopal, Unwin, White, & Alberti, 2001).
The SACSI was developed to access acute coronary syndromes symptomatology based on a literature review (DeVon & Zerwic, 2003; DeVon, Ryan, Ochs, & Shapiro, 2008). This instrument includes 20 common symptoms, location of pain or discomfort (14 items), and quality of pain or discomfort (14 items) associated with acute coronary syndromes (DeVon et al., 2008). The psychometric properties of this instrument have been tested in American and Chinese populations with good validity and reliability (DeVon & Zerwic, 2003; Li, Lee, & Yu, 2014). However, in study for Chinese population, they reported headaches, lower abdominal pain, and tearing of chest, which were not listed in SACSI. Until now, only few researchers have used the SACSI to assess cardiovascular symptoms in women. Three studies showed that women reported fewer typical symptoms than men (DeVon & Zerwic, 2003; DeVon et al., 2008; Li et al., 2014). However, this instrument does not account for differences by ethnicity/race.
The MAPMISS was developed using inductive methods and four studies were conducted to identify women’s experiences and perceptions of symptoms associated with myocardial infarction (MI) (McSweeney, O’Sullivan, Cody, & Crane, 2004). The reliability of MAPMISS was tested among White and African American women with or without chronic heart disease (CHD) (McSweeney et al., 2013). Although the MAPMISS was developed for use with an ethnically diverse population, the study sample consisted primarily of white women, African and Hispanic women of various age (McSweeney et al., 2004; McSweeney, O’Sullivan, et al., 2010; McSweeney et al., 2013).
We designed the Cardiovascular Symptom Index for Midlife Women (CSIMW) for a multiethnic/racial sample of midlife women and intended to assess the cardiovascular symptoms during menopausal transition. It was based on the Midlife Women’s Symptoms Index (MSI) (Im, 2006), and 25 of 73 items were selected based on a literature review cardiovascular symptoms of multiethnic/racial midlife women (Khan et al., 2013; Kikidis, Tsioufis, Papanikolaou, Zerva, & Hantzakos, 2014; Mayo Clinic, 2011; National Heart, Lung, and Blood Institute, 2015; Polunina et al., 2011). However, although the CSIMW has strengths for measuring cardiovascular symptoms among multiethnic/racial women, little information is available on its validity and reliability among those groups. Therefore, this study was conducted to evaluate the validity and reliability of the CSIMW in four major ethnic/racial groups of midlife women in the U.S.: whites, Asian-Americans, Hispanics, and African-Americans. The specific aims of this study were to evaluate (a) construct validity of the CSIMW using factor analysis, (b) reliability (internal consistency, inter-item correlation, and item-total correlations) and (c) convergent validity using intercorrelations of subscales with each other and with the total score.
Methods
Design
This study involved a secondary analysis of data from two internet surveys. The first study aimed to explore menopausal symptoms in multiethnic/racial groups (Im, Lee, Chee, Brown, & Dormire, 2010) and the second aimed to explore ethnic/racial differences in attitudes towards physical activity among four ethnic/racial groups of midlife women (Im et al., 2012). The institutional review boards with which the researchers were affiliated approved those studies.
Participants
Two original studies recruited participants through the Internet. We announced the study on internet communities/websites for midlife women. When the participants accessed the website developed by the research team, they were provided with information on the study and agreed to participate. The women were eligible for inclusion if they were aged between 40 and 60, could read and write English, and were Hispanic, Asian-American, African-American, or white, as self-reported ethnic/racial identity. In total, we recruited 512 participants for the first study, and 542 for the second study, after excluding the entries missing more than 10% of the data. In total, we used the data from 1,054 women for this analysis. The sample size of 1,054 participants was adequate for this analysis of psychometric properties. To run an exploratory factor analysis, the sample size met the previously suggested ratio between participants and variables of 5:1 (Kivinen, Halonen, Eronen, & Nissinen, 1998). In the Pearson correlation analysis used to test for convergent validity, a moderate correlation between variables is considered meaningful. Thus, a sample of 111 analyzable subjects provided 95% power to discover a correlation statistically different from zero at the 0.05 significance level (Cohen, Cohen, West, & Aiken, 2003).
Measures
Demographic and health-related characteristics
Questions on age, education, marital status, employment status, family income, and country of birth were used to assess demographic characteristics. Questions on diagnosed disease(s), self-reported weight, and height were used to assess health-related characteristics. Current medical history was asked using one item asking if the respondent had any diagnosed disease(s). Using an open-ended question, respondents were asked to specify the diagnosis of the disease. We calculated the BMI using weight and height and categorized as normal (≤24.9), overweight (25–29.9), or obese (≥30 kg/m2), according to the World Health Organization (WHO) definitions (World Health Organization, 2000). Menopause status was determined using questions on the last menstrual cycle, menstrual regularity, and menstrual flow. We categorized the menopause status as premenopausal, early or late perimenopausal, or postmenopausal stages.
Cardiovascular symptom index for midlife women (CSIMW)
To measure cardiovascular symptoms, the CSIMW was adapted from the MSI (Im, 2006). The CSIMW includes only the 25 questions on cardiovascular symptoms from the MSI. We choose these questions based on an extensive literature review on midlife women’s symptoms associated with cardiovascular diseases (Khan et al., 2013; Kikidis et al., 2014; Mayo Clinic, 2011; National Heart, Lung, & & Blood Institute, 2015; Polunina et al., 2011). Each item of the CSIMW includes (a) a symptom frequency subscale with dichotomous items (1 = yes, 0 = no) and (b) a symptom severity sub-scale with 6-point Likert scale items (0 = no symptom, 5 = extremely). The total numbers and the total severity scores of cardiovascular symptoms can be calculated by adding the frequency of all items (range, 0–25) and by adding the severity of the 25 items (range, 0–125). In this study, the Cronbach’s α was 0.87 for the symptom frequency subscale, and 0.89 for the symptom severity subscale.
Data analysis
Data were analyzed using the SPSS software (ver. 23.0). To describe the demographic and health-related characteristics of participants, we used descriptive statistics, including frequency, percentage, mean, and standard deviation.
Construct validity was assessed by factor analysis. Because exploratory factor analysis was not completed as part of the original study to develop CSINW, in this study, principal component factor analyses with varimax rotation was performed. The number of factors retained was determined by the eigenvalue (EV) being >1.0. Items were retained if they had an item-factor loading of at least 0.30 (Kline, 1994; Tabachnick & Fidell, 2007). To examine the replicability of these findings across multiethnic/racial groups, we performed principal component factor analysis with varimax rotation for each ethnic/racial group. The adequacy of sampling for factor analysis was calculated by the Kaiser-Meyer-Olkain (KMO) statistic, with values ranging from 0 to 1.0. KMO values greater than 0.5 indicated acceptable sampling adequacy (Kaiser, 1974). To examine convergent validity, we calculated intercorrelation coefficients of subscales with each other and with the total CSIMW score.
Internal consistency reliability was assessed with the Kuder-Richardson formula 20 (KR 20) for CSIMW’s prevalence part or Cronbach’s α for CSIMW’s severity part. KR20 and Cronbach’s α values >0.70 were considered sufficient (Streiner & Norman, 2003). Additionally, we used item analyses to check inter-item correlations and item-total correlations. To assess inter-item correlations, we used standards suggested by Briggs and Cheek (1986). Accordingly, the mean inter-item should fall with an optimal range between 0.20 and 0.50 but should not be <0.15 (Briggs & Cheek, 1986; Taylor, Bagby, & Parker, 2003). Mean inter-items >0.20 were considered a sufficient level of internal consistency or reliablity.
Results
Demographic and health-related characteristics
Table 1 shows demographic and health-related characteristics of the participants. The participants’ average age was 48.97 years old (SD = 5.69) and 86.6% had higher education than high school. About two-thirds were married or partnered, and employed. About 56% reported that, with their family income, it was somewhat or very hard to pay for basics. Among participants, 23.1% reported they were born outside the US. The participants included 255 Hispanics, 233 non- Hispanic Asian-Americans, 250 non-Hispanic African-Americans, and 316 non-Hispanic whites. Among the participants reporting that they had more than one diagnosed disease, 17.7% had cardiovascular disease(s). About 60% were overweight or obese. The menopausal stages of the women were: 27.2% premenopausal, 23.9% early peri-menopausal, 9.2% late menopausal, and 39.7% postmenopausal.
Table 1.
Demographic and health related characteristics of the participants.
| Characteristics | Hispanic n (%) |
Asian American n (%) |
African American n (%) |
White n (%) |
Total n (%) |
|---|---|---|---|---|---|
| Education | |||||
| High school | 43 (16.9) | 30 (12.9) | 17 (6.8) | 51 (16.1) | 141 (13.4) |
| High school over | 212 (83.1) | 203 (87.1) | 233 (93.2) | 265 (83.9) | 913 (86.6) |
| Marriage | |||||
| Married/partnered | 172 (67.5) | 196 (84.1) | 134 (53.6) | 212 (67.1) | 714 (67.7) |
| Non-married/partnered | 83 (32.5) | 37 (15.9) | 116 (46.4) | 104 (32.9) | 340 (32.3) |
| Employment | |||||
| Employed | 208 (81.6) | 145 (62.2) | 209 (83.6) | 228 (72.2) | 790 (75.0) |
| Unemployed | 47 (18.4) | 88 (37.8) | 41 (16.4) | 88 (27.8) | 264 (25.0) |
| Family income | |||||
| Very hard to pay | 41 (16.1) | 20 (8.6) | 32 (12.8) | 92 (29.1) | 185 (17.6) |
| Somewhat to pay | 104 (40.8) | 88 (37.8) | 102 (40.8) | 112 (35.4) | 406 (38.5) |
| Not hard to pay | 110 (43.1) | 125 (53.6) | 116 (46.4) | 112 (35.4) | 463 (43.9) |
| County of Birth | |||||
| The United States (U.S.) | 195 (76.5) | 62 (26.6) | 244 (97.6) | 309 (97.8) | 810 (76.9) |
| Outside the U.S. | 60 (23.5) | 171 (73.4) | 6 (2.4) | 7 (2.2) | 244 (23.1) |
| Diagnosed diseases | |||||
| No | 137 (53.7) | 170 (73.0) | 126 (50.4) | 149 (47.2) | 582 (55.2) |
| Yes | 118 (46.3) | 63 (27.0) | 124 (49.6) | 167 (52.8) | 472 (44.8) |
| Cardiovascular diseases | 44 (17.3) | 30 (12.9) | 65 (26.0) | 48 (15.2) | 187 (17.7) |
| Body Mass Index (kg/m2) | |||||
| Low weight (≤19.9) | 5 (2.0) | 17 (7.3) | 1 (0.4) | 4 (1.3) | 27 (2.6) |
| Normal (20.0–24.9) | 67 (26.3) | 148 (63.5) | 58 (23.2) | 123 (38.9) | 396 (37.6) |
| Overweight (25.0–29.9) | 83 (32.5) | 54 (23.2) | 61 (24.4) | 67 (21.2) | 265 (25.1) |
| Obese (≥30) | 100 (39.2) | 14 (6.0) | 130 (52.0) | 122 (38.6) | 366 (34.7) |
| Menopause status | |||||
| Premenopausal | 71 (27.8) | 78 (33.5) | 61 (24.4) | 77 (24.4) | 287 (27.2) |
| Early peri-menopausal | 61 (23.9) | 59 (25.3) | 57 (22.8) | 75 (23.7) | 252 (23.9) |
| Late peri-menopausal | 26 (10.2) | 17 (7.3) | 20 (8.0) | 34 (10.8) | 97 (9.2) |
| Postmenopausal | 97 (38.0) | 79 (33.9) | 112 (44.8) | 130 (41.1) | 418 (39.7) |
| 255 (100.0) | 233 (100.0) | 250 (100.0) | 316 (100.0) | 1054 (100.0) | |
Construct validity
Table 2 indicates the results of the CSIMW factor analysis for all participants. The KMO value (0.9) indicated the samples for all participants were tolerable. “Nose bleeds” was excluded because its factor loading was lower than 0.30. Among all the participants, we extracted five factors: cardiorespiratory symptoms, discomfort/pain, musculoskeletal symptoms, digestive disturbance, and vasomotor symptoms.
Table 2.
Cardiovascular symptoms of the participants: Varimax rotation.
| Symptoms | Cardio-respirary symptoms | Discomfort/pain | Musculoskeleteal symptoms | Digestive disturbance | Vasomotor symptoms |
|---|---|---|---|---|---|
| Heart racing | .775 | ||||
| Palpitation | .749 | ||||
| Suffocating | .641 | ||||
| Difficulty in breathing | .606 | ||||
| Pain in heart or chest | .484 | ||||
| Dizziness | .428 | ||||
| Aches in neck and skull | .710 | ||||
| Exhausion or fatigue | .566 | ||||
| Anxious or nervuous | .510 | ||||
| Numbness or tingling sense | .344 | ||||
| Headache | .340 | ||||
| Cold hands or feet trouble | .327 | ||||
| Pain in the arms or legs | .706 | ||||
| Swollen joints | .652 | ||||
| Swollen ankles | .638 | ||||
| Back pain | .496 | ||||
| Muscle and joint stiffness and soreness (neck, shoulder) | .389 | ||||
| Twitching of face, head or shoulder | .350 | ||||
| Upset stomach or nausea | .729 | ||||
| Pain in stomach | .724 | ||||
| Feel bloating after eating | .590 | ||||
| Hot flushes | .860 | ||||
| Night sweats | .781 | ||||
| Feeling hot or cold | .744 | ||||
| Eigenvalue | 2.99 | 2.73 | 2.72 | 2.37 | 2.36 |
| Cummulative variance (%) | 12.47% | 11.37% | 11.33% | 9.89% | 9.83% |
| Total variance explained (%) | 54.88% |
The results of the CSIMW factor analysis by each ethnic/racial group were as follows. KMOs values indicated the samples for subethnic/racial groups were acceptable for factor analysis. The results of Bartlett’s tests were also significant, indicating our data were suitable for factor analysis. Among Hispanics, we extracted five factors. We excluded “Nose bleeds” because its factor loading was lower than 0.30. All items in discomfort/pain except “headache” and “back pain” among musculoskeletal symptoms had high loadings on factor 1 (EV = 8.06; 33.57% of variance explained). All items of digestive disturbance had high loadings on factor 2 (EV = 1.91; 7.97% of variance explained). All items of cardiorespiratory symptoms except “dizziness” and “twitching of the face, head or shoulder” among musculoskeletal symptoms had high loadings on factor 3 (EV = 1.61; 6.71% of variance explained). Four items (“ankles swell,” “joints swollen,” “muscle and joints stiff and sore,” and “pains in the arms or legs”) in musculoskeletal symptoms, “dizziness” in cardiorespiratory symptoms and “headache” in discomfort/pain had high loadings on factor 4 (EV = 1.25; 5.22% of variance explained). All items of vasomotor symptoms had high loadings on factor 5 (EV = 1.15; 4.77% of variance explained). These five factors explained 58.23% of the total variance.
Among Asian-Americans, we extracted seven factors. All items of discomfort/pain except “headache” showed high loadings on factor 1 (EV = 7.63; 30.54% of variance explained). All items among cardiorespiratory symptoms except “dizziness” were included in factor 2 (EV = 1.63; 6.53% of variance explained). All items of of musculoskeletal symptoms except “twitching of the face, head or shoulder” had high loadings on factor 3 (EV = 1.55; 6.22% of variance explained). All items of digestive disturbance and “headache” in discomfort/pain were included in factor 4 (EV = 1.42; 5.60% of variance explained). All items of vasomotor symptoms were included in factor 5 (EV = 1.20; 4.81% of variance explained). “Dizziness” in cardiorespiratory symptoms and “twitching of the face, head or shoulder” in musculoskeletal symptoms were included in factor 6 (EV = 1.23; 4.52% of variance explained). “Nose bleeds” was included in factor 7 (EV = 1.08; 4.32% of variance explained). These seven factors explained 62.64% of the total variance.
Among African-Americans, we extracted seven factors. All items except “twitching of the face, head or shoulder” of muscloskeletal symptoms were included in factor 1 (EV = 7.23; 28.93% of variance explained). All items of digestive disturbance and “dizziness” in cardiorespiratory symtoms were included in factor 2 (EV = 1.62; 6.47% of variance explained). All items of discomfort/pain except “cold hands or feet” were included in factor 3 (EV = 1.55; 6.21% of variance explained). All items of cardiorespiratory symptoms except “bothered by palpitations” and “twitching of the face, head or shoulder” in musculoskeletal symptoms were included in factor 4 (EV = 1.27; 5.08% of variance explained). All items of vasomotor symptoms were included in factor 5 (EV = 1.21; 4.82% of variance explained). “Bothered by palpitations” in cardiorespiratory symptoms and “cold hands or feet” in discomfort/pain were included in factor 6 (EV = 1.09; 4.37% of variance explained). “Nose bleeds” was included in factor 7 (EV = 1.04; 4.15% of variance explained). These seven factors explained 60.03% of the total variance.
Among white participants, we extracted five factors. “Nose bleeds” was excluded because its factor loading was lower than 0.30. All items of musculoskeletal symptoms except “cold hands or feet trouble” and “headaches” and “muscle and joints stiff and sore” of muscloskeletal symptoms had high loadings on factor 1 (EV = 8.35, 34.80% of variance explained). All items of cardiorespiratory symptoms had high loadings on factor 2 (EV = 1.78; 7.40% of variance explained). “Headache” of discomfort/pain, and all items of digestive disturbance were included in factor 3 (EV = 1.41, 5.89% of variance explained). All items of vasomotor symptoms had high loadings on factor 4 (EV = 1.26; 5.25% of variance explained). All items except “muscle and joints stiff and sore” of muscloskeletal symptoms and “cold hands or feet trouble” of discomfort/pain had high loadings on factor 5 (EV = 1.13; 4.69% of variance explained). These five factors explained 58.03% of the total variance.
Reliability
Table 3 summarizes KR 20 values and Cronbach’s α coefficients of the CSIMW by ethnicity/race to assess internal consistency reliability. The results indicated adequate internal consistency of the CSIMW (both the prevalence part and severity parts) for all ethnic/racial groups. Among the subscales, KR 20 values and Cronbach’s α values for cardiorespiratory symptoms, discomfort/pain, musculoskeletal symptoms, and vasomotor symptoms were acceptable, but they were not acceptable for digestive disturbances for all participants. For Hispanic women, KR 20 values of the prevalence part of discomfort/pain and digestive symptoms were not acceptable. For African-American women, KR 20 values of the prevalence part of the subscales except discomfort/pain were not acceptable. For Asian-American women, the CSIMW prevalence part and severity part of digestive disturbance and vasomotor symptoms were not acceptable. For white women, the CSIMW prevalence part and severity part of digestive disturbance and the prevalence part of musculoskeletal symptoms were not acceptable.
Table 3.
Internal consistency reliability of the CSIMW by ethnicity/race.
| Characteristics | Hispanic | Asian American | African American | White | Total | |
|---|---|---|---|---|---|---|
| Prevalence part | Total score | .89 | .87 | .87 | .89 | .89 |
| Cardiorespiratory symptoms | .77 | .70 | .68 | .76 | .74 | |
| Discomfort/pain | .67 | .71 | .70 | .71 | .70 | |
| Musculoskeletal symptoms | .76 | .71 | .68 | .68 | .71 | |
| Digestive disturbance | .66 | .55 | .57 | .63 | .62 | |
| Vasomotor symptoms | .74 | .57 | .66 | .71 | .71 | |
| Severity part | Total score | .91 | .90 | .89 | .92 | .91 |
| Cardiorespiratory symptoms | .82 | .78 | .71 | .83 | .80 | |
| Discomfort/pain | .74 | .75 | .72 | .77 | .76 | |
| Musculoskeletal symptoms | .80 | .76 | .74 | .75 | .77 | |
| Digestive disturbance | .74 | .60 | .62 | .67 | .67 | |
| Vasomotor symptoms | .81 | .68 | .74 | .84 | .81 |
Note: Values are Kuder–Richardson Formula 20 for prevalence part or Cronbach’s α for severity part.
Table 4 shows the results of item-total correlations and inter-item correlations. The item-total correlation coefficients of all items in the prevalence part were more than 0.20 for all ethnic/racial groups. Item-total correlations of Hispanics were 0.35–0.58 and the average inter-item correlation coefficients for total scores was 0.25. Among Asian-American participants, item-total correlation coefficients ranged from 0.25 to 0.58 and the average inter-item correlation coefficient for total scores was 0.22. Among African-American participants, item-total correlation coefficients ranged from 0.33 to 0.56 and the average inter-item correlation coefficient for total scores was 0.23. Item-total correlation coefficients for whites ranged from 0.38 to 0.58 and the average inter-item correlation coefficient for total scores was 0.26. The results of the item analysis on the severity part of CSIMW were similar to those on the prevalence part. The item-total correlation coefficients of all items in the severity part were > 0.20 for all ethnic/racial groups.
Table 4.
The results of item analyses by ethnicity/race.
| Characteristics | Hispanic | Asian American | African American | White | Total | |
|---|---|---|---|---|---|---|
| Prevalence part | Item-total correlation(range) | .35–.58 | .25–.58 | .33–.56 | .38–.58 | .41–.57 |
| Inter-item correlation(average) | .25 | .22 | .23 | .26 | .25 | |
| Severity part | Item-total correlation(range) | .42–.65 | .42–.65 | .36–.59 | .44–.64 | .45–.65 |
| Inter-item correlation(average) | .30 | .28 | .27 | .32 | .30 |
Note: Values are Kuder-Richardson Formula 20 for prevalence part or Cronbach’s α for severity part.
Convergent validity
Convergent validity was confirmed by intercorrelations of sub-symptoms, with each other and with the total score (Table 5). Across all ethnic/racial groups, total prevalence CSIMW scores were correlated with all subscale scores, as evidenced by r values ranging from 0.54 to 0.89. The subscales of the severity parts had positive correlation with each other for all ethnic/racial groups, ranging from 0.60 to 0.91.
Table 5.
Intercorrelations of the subdomains of the CSIMW by Ethnicity/race.
| Hispanic | ||||||
| Discomfort/Pain r(p) | Musculoskeletal symptoms r(p) | Digestive disturbance r(p) | Vasomotor symptoms r(p) | Total r(p) | ||
|
| ||||||
| Prevalence part | Cardiovascular symptoms | .55 (<.01) | .49 (<.01) | .41 (<.01) | .33 (<.01) | .75 (<.01) |
| Discomfort/pain | .67 (<.01) | .48 (<.01) | .48 (<.01) | .86 (<.01) | ||
| Musculoskeletal symptoms | .52 (<.01) | .41 (<.01) | .83 (<.01) | |||
| Digestive disturbance | .31 (<.01) | .68 (<.01) | ||||
| Vasomotor symptoms | .64 (<.01) | |||||
| Severity part | Cardiovascular symptoms | .58 (<.01) | .50 (<.01) | .44 (<.01) | .36 (<.01) | .74 (<.01) |
| Discomfort/pain | .71 (<.01) | .55 (<.01) | .52 (<.01) | .89 (<.01) | ||
| Musculoskeletal symptoms | .54 (<.01) | .44 (<.01) | .85 (<.01) | |||
| Digestive disturbance | .36 (<.01) | .71 (<.01) | ||||
| Vasomotor symptoms | .67 (<.01) | |||||
|
| ||||||
| Asian American |
||||||
| Discomfort/Pain r(p) | Musculoskeletal symptoms r(p) | Digestive disturbance r(p) | Vasomotor symptoms r(p) | Total r(p) | ||
|
| ||||||
| Prevalence part | Cardiovascular symptoms | .56 (<.01) | .52 (<.01) | .47 (<.01) | .30 (<.01) | .78 (<.01) |
| Discomfort/pain | .62 (<.01) | .45 (<.01) | .35 (<.01) | .86 (<.01) | ||
| Musculoskeletal symptoms | .44 (<.01) | .29 (<.01) | .82 (<.01) | |||
| Digestive disturbance | .21 (<.01) | .65 (<.01) | ||||
| Vasomotor symptoms | .54 (<.01) | |||||
| Severity part | Cardiovascular symptoms | .62 (<.01) | .60 (<.01) | .56 (<.01) | .37 (<.01) | .81 (<.01) |
| Discomfort/pain | .68 (<.01) | .52 (<.01) | .47 (<.01) | .89 (<.01) | ||
| Musculoskeletal symptoms | .46 (<.01) | .34 (<.01) | .83 (<.01) | |||
| Digestive disturbance | .29 (<.01) | .69 (<.01) | ||||
| Vasomotor symptoms | .60 (<.01) | |||||
|
| ||||||
| African American |
||||||
| Discomfort/Pain r(p) | Musculoskeletal symptoms r(p) | Digestive disturbance r(p) | Vasomotor symptoms r(p) | Total r(p) | ||
|
| ||||||
| Prevalence part | Cardiovascular symptoms | .53 (<.01) | .54 (<.01) | .56 (<.01) | .35 (<.01) | .78 (<.01) |
| Discomfort/pain | .60 (<.01) | .55 (<.01) | .40 (<.01) | .85 (<.01) | ||
| Musculoskeletal symptoms | .50 (<.01) | .36 (<.01) | .81 (<.01) | |||
| Digestive disturbance | .37 (<.01) | .74 (<.01) | ||||
| Vasomotor symptoms | .61 (<.01) | |||||
| Severity part | Cardiovascular symptoms | .59 (<.01) | .58 (<.01) | .55 (<.01) | .46 (<.01) | .80 (<.01) |
| Discomfort/pain | .64 (<.01) | .52 (<.01) | .47 (<.01) | .86 (<.01) | ||
| Musculoskeletal symptoms | .50 (<.01) | .42 (<.01) | .83 (<.01) | |||
| Digestive disturbance | .41 (<.01) | .72 (<.01) | ||||
| Vasomotor symptoms | .69 (<.01) | |||||
|
| ||||||
| White |
||||||
| Discomfort/Pain r(p) | Musculoskeletal symptoms r(p) | Digestive disturbance r(p) | Vasomotor symptoms r(p) | Total r(p) | ||
|
| ||||||
| Prevalence part | Cardiovascular symptoms | .65 (<.01) | .56 (<.01) | .50 (<.01) | .38 (<.01) | .82 (<.01) |
| Discomfort/pain | .68 (<.01) | .55 (<.01) | .46 (<.01) | .89 (<.01) | ||
| Digestive & circulation disturbance | .48 (<.01) | .38 (<.01) | .82 (<.01) | |||
| Musculoskeletal symptoms | .37 (<.01) | .71 (<.01) | ||||
| Vasomotor symptoms | .63 (<.01) | |||||
| Severity part | Cardiovascular symptoms | .64 (<.01) | .60 (<.01) | .49 (<.01) | .39 (<.01) | .79 (<.01) |
| Discomfort/pain | .73 (<.01) | .61 (<.01) | .53 (<.01) | .91 (<.01) | ||
| Digestive & circulation disturbance | .49 (<.01) | .44 (<.01) | .85 (<.01) | |||
| Musculoskeletal symptoms | .41 (<.01) | .71 (<.01) | ||||
| Vasomotor symptoms | .68 (<.01) | |||||
|
| ||||||
| Total |
||||||
| Discomfort/Pain r(p) | Musculoskeletal symptoms r(p) | Digestive disturbance r(p) | Vasomotor symptoms r(p) | Total r(p) | ||
|
| ||||||
| Prevalence part | Cardiovascular symptoms | .59 (<.01) | .54 (<.01) | .49 (<.01) | .37 (<.01) | .79 (<.01) |
| Discomfort/pain | .65 (<.01) | .53 (<.01) | .44 (<.01) | .87 (<.01) | ||
| Musculoskeletal symptoms | .50 (<.01) | .38 (<.01) | .82 (<.01) | |||
| Digestive disturbance | .34 (<.01) | .70 (<.01) | ||||
| Vasomotor symptoms | .63 (<.01) | |||||
| Severity part | Cardiovascular symptoms | .62 (<.01) | .57 (<.01) | .50 (<.01) | .41 (<.01) | .79 (<.01) |
| Discomfort/pain | .71 (<.01) | .58 (<.01) | .52 (<.01) | .89 (<.01) | ||
| Musculoskeletal symptoms | .51 (<.01) | .44 (<.01) | .84 (<.01) | |||
| Digestive disturbance | .39 (<.01) | .71 (<.01) | ||||
| Vasomotor symptoms | .68 (<.01) | |||||
Discussion
Examining the psychometric equivalence of CSIMW, a predicting measure of cardiovacular disaease, across ethnic/racial groups was a necessary step to accurately measure and examine possible differences between the groups. The results of this study indicated the CSIMW had similar psychometric properites across ethnic/racial groups in several ways. First, the results of the factors analysis revealed the emergence of various factors across groups. CSIMW includes five sets of symptoms: cardiorespiratory symptoms, discomfort/pain, musculoskeletal symptoms, digestive disturbances, and vasomotor symptoms. All items of vasomotor symptoms and digestive disturbances were extracted in the same way across all ethnic/racial groups. However, differences were seen in discomport/pain, musculoskeletal symptoms, and cardiorespiratory symptoms among the ethnic/racial groups. More than five factors were extracted in Asian-Americans and African-Americans. Items included in the same factors were related to each other. Women experience symptoms that belong to the same factor at the same time. Thus, these results reflect ethnic/racial differences in cardiovascular symptoms. These findings were similar to those of previous studies (McSweeney, Cleves, Zhao, Lefler, & Yang, 2010; Teoh, Lalondrelle, Roughton, Grocott-Mason, & Dubrey, 2005). In their study on acute and prodromal symptoms of myocardial infarction using MAPMISS (McSweeney, Cleves, et al., 2010), it was reported that white and Hispanic women were more likely to be asymptomatic groups than African Americans, but African American women were more likely to experience multiple distressing symptoms, such as shortness of breath, arms weak/heavy, hands/arms tingling, sleep disturbances, and unusual fatigue. Teoh et al. (2005) reported that Asian patients tended to report a higher intensity of pain, over a greater area of their body, and more frequent discomfort over the rear of their upper thorax than Caucasian patients in long-term prospective survey. Women younger than 55 years reported a great number of non-chest pain symptoms than men in acute coronary syndrome presentation (Khan et al., 2013). Ethnic/racial difference in cardiovascular symptoms among midlife women depended on type of symptoms (McSweeney, Cleves, et al., 2010; McSweeney, O’Sullivan, et al., 2010; Im, Ham, Chee, & Chee, 2015). Therefore, the CSIMW can be useful to assess cardiovascular symptom among multiethic/raical midlife women.
Another possible explanation for the results here is the high correlation between menopause and cardiovascular symptoms. The participants of this study were mainly of healthy women in the transition of menopause. Subsequently, their menopausal transition may have contributed to the symptoms reported among the ethnic/racial groups. Emerging evidence supports that women with menopausal vasomotor symptoms have increased cardiovascular disease risks, as measured by surrogate markers (Szmuilowicz et al., 2011). Specifically, Szmuilowicz et al. (2011) reported that the predictive value of menopausal vasomotor symptoms for clinical cardiovascular diesease events may vary with the onset of menopausal vasomotor symptoms at different stages of menopause. Richard-Davis and Wellons (2013) also said that there were complex ethnic/racial differences that varied with menopausal state, and age at that menopausal state. In this study, the unequal distribution of menopausal stages across the ethnic/racial groups may have affected the emergence of various symptoms across the groups. Thus, the results of the factor analyses indicated that the factor structure was changed by target ethnic/racial group or by the stage of menopause the participants were experiencing.
The “nose bleeds” item had a low factor loading value among all participants, especially Hispanic and white women, indicating that this item may have no relationship with other cardiovascular symptoms. Because epistaxis is a common symptom of diverse conditions, Hispanic and white women may report that “nose bleed” symptoms occurred, but for other reasons. Several studies have reported that epistaxis may be a presenting sign of hypertension (Page, Biet, Liabeuf, Strunski, & Fournier, 2011), while others have reported no relationship with hypertension (Sarhan & Algamal, 2015). However, it is possible that hypertension causes atherosclerotic nasal vascular changes that predispose hyptertensives to increased susceptibility to epistaxis (Isezuo et al., 2008). Thus, long-lasting hypertension may contribute to epistaxis. Thus, “nose bleed,” in terms of frequency, intensity, and duration should be examined carefully.
The findings of the reliablity tests showed adequate internal consistency of total scores for all ethnic/racial groups, supporting the use of CSIMW in the various ethnic/racial groups. However, some subscales were below the criterion of an α value of at least 0.70. A possible reason for the decreased values of internal consistency coefficients in the subscale could be heterogeneity of configuration questionnaires and the low response rate (McSweeney et al., 2004). The KR 20 values of “digestive disturbance” were from 0.55 to 0.74. This was lower than those of other domains. The KR 20 of the prevalence part of “discomfort/pain” for Hispanic women was 0.67. There are some reports that women had the fewest and least severe symptoms during the prodromal phase (McSweeney et al., 2003). A lower response rate on a particular item could result in unstable coefficients (McSweeney et al., 2004). Furthermore, several researchers reported significant ethnic/racial differences in cardiovascular symptoms among multiethnic/racial women (Eastwood et al., 2013; McSweeney, O’Sullivan, et al., 2010; Meshack et al., 1998). African American women were more likely to report more digestive disturbance symptoms, such as loss of appetite, indigestion, and vomiting, than white women (Eastwood et al., 2013; McSweeney, O’Sullivan, et al., 2010), while Hispanic women reported more “discomfort/pain” symptoms, such as upper back pain and anxiety than whites (McSweeney, O’Sullivan, et al., 2010; Meshack et al., 1998). Thus, those domains may include some ethnic/raical-specific symptoms and this could contribute to the lower α coefficients. Another possible explanation for the decreased values of internal consistency coefficients in subscales such as “vasomotor symptoms,” and “digestive disturbance” may be due to the small number of items (3 items) (Hayes, 2008). This instrument was developed based on a literature review for ethnically/racially diverse midlife women. Some of questions and wording, such as severity and frequency, and the expression of symptoms may be interpretely differently by those of different cultural backgrounds. Thus, these questions may cause low reliability (Meshkani & Hossein, 2005). Especially, KR 20 values for all subscales in the prevalence part for African-Americans were lower than the others. This low reliablility may be due to African-American women’s symptoms. McSweeney, O’Sullivan, et al. (2010) reported that African women had higher frequencies and intensities of both prodromal and acute MI symptoms versus Caucasian women. The results of this study also supported the possibility that African-American women had a different variety of cardiovascular symptoms than the other ethnic/racial groups.
Finally, in the convergent validity tests, all ethnic/racial groups showed similar patterns. All subscales converged into total prevalence and severity scores and subscales were highly correlated with each other. This provides evidence that there was no difference in convergent validity among the four ethnic/racial groups. Overall, the CSIMW proved to be a reliable instrument and was reasonably accurate in assesing cardiovascular symptoms for multiethnic/racial groups. The CSIMW has great potential as a screening instrument that is easy to administer and can be used in multiethnic/racial groups of midlife women.
This study has some limitations. First, the sample was recruited from the Internet. The generalizability may be limited to multiethnic/racial midlife women with similar characteristics to the study population and may not extend to all multiethnic/racial midlife women in the US. Second, this questionnaire was adapted from the MSI and the items were selected based on a literature review. Further observational studies of ethnic/racial -specific cardiovascular symptom patterns will help to validate the questionnaire. Further studies are needed to determine the predictive capacity of the cardiovascular symptoms included in the instrument. This will help health care providers determine which women should be referred for cardiovascular diagnostic evaluations.
Finally, a lack of factorial invariance of CSIMW was found in this study, which may lead us to biased mean values in interpreting the study findings. Thus, further psychometric and invariance testing is recommended in multiethnic/racial midlife women.
Conclusions
Despite this study’s limitations, these findings indicate that the CSIMW is useful for assessing cardiovascular symptoms in women, although the symptom contruct of CSIMW can be considered to vary by ethnicity/race. Based on this study, we propose the following implications for further studies and practice. First, additional studies with diverse groups of multiethnic/racial midlife women are needed to confirm CSIMW’s psychometric properties. Second, we suggest that the item “nose bleed” be revised to ask for more information. Finally, health care professionals need to pay close attention to ethnic/racial differences in cardiovascular symptoms. Future studies will be needed to identify ethnic/racial-specific cardiovascular symptom patterns and to estabilish the predictive validity of CSIMW scores. Knowledge regarding cardiovascular symptom presentation in multiethnic/racial women can improve the identification of women who are at high risk for developing CHD and assist in providing timely treatment and, thus, improve health care outcomes.
Funding
The original 2 studies that provided the data for this secondary analysis were funded by the National Institutes of Health (NIH/NINR/NIA and NIH/NINR/NHLBI) (R01NR008926 and R01NR010568). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Disclosure statement
No potential conflict of interest was reported by the author(s).
References
- Briggs SR, & Cheek JM (1986). The role of factor analysis in the development and evaluation of personality-scales. Journal of Personality, 54(1), 106–148. doi: 10.1111/j.1467-6494.1986.tb00391.x [DOI] [Google Scholar]
- Cohen J, Cohen P, West SG, & Aiken LS (2003). Applied multiple regression/correlation analysis for the behavioral science (3rd ed.). Mahwah, NJ: Lawrence Erlbaum. [Google Scholar]
- DeVon HA, & Zerwic JJ (2003). The symptoms of unstable angina: Do women and men differ. Nursing Research, 52(2), 108–118. doi: 10.1097/00006199-200303000-00007 [DOI] [PubMed] [Google Scholar]
- DeVon HA, Ryan CJ, Ochs AL, & Shapiro M (2008). Symptoms across the continuum of acute coronary syndromes: Differences between women and men. American Journal of Critical Care, 17(1), 14–24. [PMC free article] [PubMed] [Google Scholar]
- Eastwood J-A, Johnson BD, Rutledge T, Bittner V, Whittaker KS, Krantz DS, … Bairey Merz CN (2013). Anginal symptoms, coronary artery disease, and adverse outcomes in black and white women: The NHLBI-sponsored women’s ischemia syndrome evaluation (WISE) study. Journal of Women’s Health, 22(9), 724–733. doi: 10.1089/jwh.2012.4031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fischbacher CM, Bhopal R, Unwin N, White M, & Alberti KG (2001). The performance of the Rose angina questionnaire in South Asian and European origin populations: A comparative study in Newcastle, UK. International Journal of Epidemiology, 30(5), 1009–1016. doi: 10.1093/ije/30.5.1009 [DOI] [PubMed] [Google Scholar]
- Harris RB, & Weissfeld LA (1991). Gender differences in the reliability of reporting symptoms of angina pectoris. Journal of Clinical Epidemiology, 44(10), 1071–1078. [DOI] [PubMed] [Google Scholar]
- Hayes BE (2008). Measuring customer satisfaction and loyalty: Survey design, use, and statistical analysis method (3rd ed.). Milwaukee, WI: American Society for Quality. [Google Scholar]
- Im EO (2006). The midlife women’s symptom index (MSI). Health Care for Women International, 27(3), 268–287. doi: 10.1080/07399330500506600 [DOI] [PubMed] [Google Scholar]
- Im EO, Chang SJ, Ko Y, Chee W, Stuifbergen A, & Walker L (2012). A national internet survey on midlife women’s attitudes toward physical activity. Nursing Research, 61(5), 342–352. doi: 10.1097/NNR.0b013e31825da85a [DOI] [PubMed] [Google Scholar]
- Im EO, Ham OK, Chee E, & Chee W (2015). Racial/ethnic difference in cardiovascular symptoms in four major racial/ethnic groups of midlife women: A secondary analysis. Women & Health, 55(5), 525–547. doi: 10.1080/03630242.2015.1022813 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Im EO, Lee B, Chee W, Brown A, & Dormire S (2010). Menopausal symptoms among four major ethnic groups in the United States. Western Journal of Nursing Research, 32(4), 540–565. doi: 10.1177/0193945909354343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Isezuo SA, Segun-Busari S, Ezunu E, Yakubu A, Iseh K, Legbo J, … Ologe FE (2008). Relationship between epistaxis and hyptertension: A study of patients seen in the emergency units of two tertiary health institutions in Nigeria. Nigerian Journal of Clinical Practice, 11(4), 379–382. [PubMed] [Google Scholar]
- Kaiser HF (1974). An index of factorial simplicity. Psychometrika, 39(1), 31–36. doi: 10.1007/BF02291575 [DOI] [Google Scholar]
- Khan NA, Daskalopoulou SS, Karp I, Eisenberg MJ, Pelletier R, Tsadok MA, … Pilote L (2013). Sex differences in acute coronary syndrome symptom presentation in young patients. JAMA Internal Medicine, 173(20), 1863–1871. doi: 10.1001/jamainternmed.2013.10149 [DOI] [PubMed] [Google Scholar]
- Kikidis D, Tsioufis K, Papanikolaou V, Zerva K, & Hantzakos A (2014). Is epistaxis associated with arterial hypertension? A systematic review of the literature. European Archives of Oto-Rhino-Laryngology, 271(2), 237–243. doi: 10.1007/s00405-013-2450-z [DOI] [PubMed] [Google Scholar]
- Kivinen P, Halonen P, Eronen M, & Nissinen A (1998). Self-rated health, physician-rated health and associated factors among elderly men: The Finnish cohorts of the Seven Countries Study. Age and Ageing, 27(1), 41–47. doi: 10.1093/ageing/27.1.41 [DOI] [PubMed] [Google Scholar]
- Kline P (1994). An easy guide to factor analysis. London: Routledge. [Google Scholar]
- Li PW, Lee DT, & Yu DS (2014). Psychometric evaluation of the symptoms of acute coronary syndromes inventory in Chinese patients with acute coronary syndrome. European Journal of Cardiovascular Nursing, 13(4), 295–303. doi: 10.1177/1474515113509558 [DOI] [PubMed] [Google Scholar]
- Mayo Clinic. (2011, March 20). Health Information: Heart Disease. Retrieved from http://www.mayoclinic.org/diseases-conditions/heart-disease/home/ovc-20341555.
- McSweeney JC, Cleves MA, Zhao W, Lefler LL, & Yang S (2010). Cluster analysis of women’s prodromal and acute myocardial infarction symptoms by race and other characteristics. The Journal of Cardiovascular Nursing, 25(4), 311–322. doi: 10.1097/JCN.0b013e3181cfba15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McSweeney JC, Cleves MA, Fischer EP, Rojo MO, Armbya N, & Moser DK (2013). Reliability of the McSweeney acute and prodromal myocardial infarction symptom survey among black and white women. European Journal of Cardiovascular Nursing, 12(4), 360–367. doi: 10.1177/1474515112459989 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McSweeney JC, Cody M, O’Sullivan P, Elberson K, Moser DK, & Garvin BJ (2003). Women’s early warning symptoms of acute myocardial infarction. Circulation, 108(21), 2619–2623. doi: 10.1161/01.CIR.0000097116.29625.7C [DOI] [PubMed] [Google Scholar]
- McSweeney JC, O’Sullivan P, Cleves MA, Lefler LL, Cody M, Moser DK, … Zhao W (2010). Racial differences in women’s prodromal and acute symptoms of myocardial infarction. American Journal of Critical Care, 19(1), 63–73. doi: 10.4037/ajcc2010372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McSweeney JC, O’Sullivan P, Cody M, & Crane PB (2004). Development of the McSweeney acute and prodromal myocardial infarction symptom survey. The Journal of Cardiovascular Nursing, 19(1), 58–67. doi: 10.1097/00005082-200401000-00010 [DOI] [PubMed] [Google Scholar]
- Meshack AF, Goff DC, Chan W, Ramsey D, Linares A, Reyna R, … Pandey D (1998). Comparison of reported symptoms of acute myocardial infarction in Mexican Americans Versus non-Hispanic Whites (The corpus Christ heart project). The American Journal of Cardiology, 82(11), 1329–1332. doi: 10.1016/S0002-9149(98)00636-5 [DOI] [PubMed] [Google Scholar]
- Meshkani Z, & Hossein AF (2005). Multivariate analysis of factors: Influencing reliability of teacher made tests. Journal of Medical Education, 6(2), 149–152. [Google Scholar]
- National Heart, Lung, and Blood Institute. (2015, March 21). What are the signs and symptoms of heart disease? Retrieved from http://www.nhlbi.nih.gov/health/health-topics/topics/hdw/signs.html.
- Nicholson A, White IR, Macfarlane P, Brunner E, & Marmot M (1999). Rose Questionnaire angina in younger men and women: Gender differences in the relationship to cardiovascular risk factors and other reported symptoms. Journal of Clinical Epidemiology, 52(4), 337–346. doi: 10.1016/S0895-4356(99)00007-4 [DOI] [PubMed] [Google Scholar]
- Oei HH, Vliegenthart R, Deckers JW, Hofman A, Oudkerk M, & Witteman JC (2004). The association of rose questionnaire angina pectoris and coronary calcification in a general population: The Rotterdam Coronary Calcification Study. Annals of Epidemiology, 14(6), 431–436. doi: 10.1016/j.annepidem.2003.09.009 [DOI] [PubMed] [Google Scholar]
- Page C, Biet A, Liabeuf S, Strunski V, & Fournier A (2011). Serious spontaneous epistaxis and hypertension in hospitalized patients. European Archives of Oto-Rhino-Laryngology, 268(12), 1749–1753. doi: 10.1007/s00405-011-1659-y [DOI] [PubMed] [Google Scholar]
- Polunina A, Gugleta K, Kochkorov A, Katamay R, Flammer J, & Orgül S (2011). Relationship between peripheral blood flow in extremities and choroidal circulation. Klinische Monatsblätter Für Augenheilkunde, 228(4), 302–305. doi: 10.1055/s-0031-1273211 [DOI] [PubMed] [Google Scholar]
- Richard-Davis G, & Wellons M (2013). Racial and ethnic differences in the physiology and clinical symptoms of menopause. Seminars in Reproductive Medicine, 31, 380–386. doi: 10.1055/s-0033-1348897 [DOI] [PubMed] [Google Scholar]
- Rose G, McCartney P, & Reid DD (1977). Self-administration of a questionnaire on chest pain and intermittent claudication. Journal of Epidemiology & Community Health, 31, 42–48. doi: 10.1136/jech.31.1.42 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rosenfeld AG (2001). Women’s risk of decision delay in acute myocardial infarction: Implications for research and practice. AACN Clinical Issues: Advanced Practice in Acute & Critical Care, 12(1), 29–39. doi: 10.1097/00044067-200102000-00005 [DOI] [PubMed] [Google Scholar]
- Sarhan NA, & Algamal AM (2015). Relationship between epistxis and hypertension: A cause and effect or coincidence? Journal of the Saudi Heart Association, 27(2), 79–84. doi: 10.1016/j.jsha.2014.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sorlie PD, Cooper L, Schreiner PJ, Rosamond W, & Szklo M (1996). Repeatability and validity of the rose questionnaire for angina pectoris in the atherosclerosis risk in communities study. Journal of Clinical Epidemiology, 49(7), 719–725. doi: 10.1016/0895-4356(96)00022-4 [DOI] [PubMed] [Google Scholar]
- Streiner D, & Norman G (2003). Health measurement scales, practical guide to their development and use (3rd ed.). Oxford: Oxford University Press. [Google Scholar]
- Szmuilowicz ED, Manson JE, Rossouw JE, Howard BV, Margolis KL, Greep NC, … Seely EW (2011). Vasomotor symptoms and cardiovascular events in postmenopausal women. Menopause, 18(6), 603–610. doi: 10.1097/gme.0b013e3182014849 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tabachnick BG, & Fidell LS (2007). Using multivariate statistics (5th ed.). Boston, MA: Allyn & Bacon. [Google Scholar]
- Taylor GJ, Bagby RM, & Parker JD (2003). The 20-item Toronto Alexithymia Scale. IV. reliability and factorial validity in different languages and cultures. Journal of Psychosomatic Research, 55(3), 277–283. doi: 10.1016/S0022-3999(02)00601-3 [DOI] [PubMed] [Google Scholar]
- Teoh M, Lalondrelle S, Roughton M, Grocott-Mason R, & Dubrey SW (2005). Acute coronary syndromes and their presentation in Asian and Caucasian patients in Britain. Heart, 93(2), 183–188. doi: 10.1136/hrt.2006.091900 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thurston RC, Sutton-Tyrrell K, Everson-Rose SA, Hess R, & Matthews KA (2008). Hot flashes and subclinical cardiovascular disease: Findings from the Study of Women’s Health across the Nation Heart Study. Circulation, 118(12), 1234–1240. doi: 10.1161/CIRCULATIONAHA.108.776823 [DOI] [PMC free article] [PubMed] [Google Scholar]
- World Health Organization. (2000). Obesity: Preventing and managing the global epidemic. Retrieved from http://www.who.int/nutrition/publications/obesity/WHO_TRS_894/en/. [PubMed]